Gut Liver. 2026 Jun 22. doi: 10.5009/gnl250626. Online ahead of print.
ABSTRACT
BACKGROUND/AIMS: The metabolic dysfunction-associated fibrosis 5 (MAF-5) tool has been proposed for identifying individuals at high risk of liver fibrosis, but its ability to predict liver-related events (LREs) remains unknown. We aimed to evaluate the ability of MAF-5 to predict LREs and assess whether modifications could enhance its predictive capacity.
METHODS: A retrospective cohort of 62,625 adults without cancer, organ transplantation, chronic viral hepatitis, or heavy alcohol intake was followed for LREs (hepatocellular carcinoma and/or liver cirrhosis complications). The MAF-5 score was calculated and compared with other non-invasive liver fibrosis biomarkers. We also assessed whether modifying the MAF-5 score could improve LRE prediction.
RESULTS: During a median follow-up of 11.2 years, 147 patients developed LREs. The MAF-5 scores were used to stratify participants by LRE risk into low-, intermediate-, and high-risk categories, with incidence rates of 0.108, 0.576, and 1.520 cases per 1,000 person-years, respectively. Age was identified as an independent risk factor for LREs, and therefore, we developed an age-modified MAF-5 (aMAF-5) score, which showed improved performance (C-index: 0.870 vs 0.818; integrated area under the curve: 0.858 vs 0.784). Within the same MAF-5 category, LRE risk varied according to the aMAF-5 score. The use of “either positive” criteria improved sensitivity (from 0.412 to 0.765) and decreased specificity (from 0.944 to 0.830), while “both positive” criteria improved specificity and reduced sensitivity. Both scores performed well regardless of age, sex, or metabolic syndrome status.
CONCLUSIONS: The MAF-5 score allows effective stratification of LRE risk. The aMAF-5 score further improves risk stratification. These scores identify individuals at risk of LREs who may benefit from enhanced surveillance.
PMID:42325012 | DOI:10.5009/gnl250626
